Reverb in Ableton - Understanding Each Parameter

Today, I am going to look at Ableton Live 8's built in reverb. This is not so much a tutorial, but a way for myself to look at it better. Like I mentioned last week, i'm hitting the books hard for the next few weeks, and I felt like I really didnt understand all of the reverb options. So, after reading a few books, and playing around with it a while...I wanted to note everything piece by piece to help me remember. I toss on a spectrum to give some visual feedback too.

Reverberation is a natural occurrence that happens after a sound leaves a speaker, and before it reaches your ear. It is not something that occurs in synthesis, and therefor, if we want to create it...we must do it artificially. When sound leaves a source, some of the vibrations (if you are close to the source) will reach your ears directly. The other sounds will bounce off other surfaces like walls, the ground, buildings, people, your own body, etc...and then reach your ear. This is what causes reverb. Our brains are actually so used to this occurrence, that we are able to determine our location size (such as a room, or in a dense forest) without actually using our eyes or other senses. Our brains actual enjoy this sound, and can give the deception of distance to sound. These sounds bounce off other objects like echoes...but so close in time to each other that they cannot be heard distinctly themselves. As you were to go farther away from a large reflective surface, the time between echoes gets larger as the sound must travel further. Eventually, you will be able to hear distance echoes in a delay fashion, just like when you yell into a valley.



Now, what is even more interesting about reverb is how a wave reflects differently off of certain objects. Since a surface has a direct frequency response, depending on which frequencies the surface absorbs, it will reflect different frequencies. This is why you put padding on walls of a studio because the softer surface will absorb more of the high frequencies making less natural reverb in the room. If you were playing in a big church or something, it would be the opposite...whereas the hard walls reflect the high frequencies.

By using reverb correctly, you can trick the ears into perceiving audio as further away without actual lowering your volume fader. Half of music is the science, the other half is how our brain is tricked into perceiving it. We must understand both to understand reverb especially.

So lets begin. For the purpose of this tutorial, I have put in a snare hit so we can easily see the spectrum change with each snapshot I take.

Here we are with the snare sample running, with the reverb off. There is a nice little peak around 1khz. Its easier to see the dry wet I will show in a minute if we switch the spectrum to "bins" instead of "lines".

REFER TO THIS AS NORMAL

Now, when we put the reverb up to full blast on its normal settings (just drag and drop), you will see a db decrease, and also muddiness in the valleys of the frequencies. Also, if I were to show you a few snapshots of the same reverb in action, you would notice small, unpredictable changes in different frequencies of the waveform.

You notice the longer valleys in the high frequencies are mostly gone with the reverb up. That is what reverb is doing to the frequency spectrum.

Now, we have a lot of control however on what exactly it is doing to these frequencies. Lets go through all the presets on the reverb. (left to right)

First, the whole reverb comes through the filters. You have a high cut and low cut. You can shut them off to save CPU power. Below them are your normal frequency and bandwidth options that are on normal filters. This can be really important if you actually want to kinda of "tune" your 1st reverb reflection to match other elements, or the element that you are putting the reverb on. I usually never start without touching this with a spectrum first.

Heres one with the highs standing out -

Here is one with the lows sounding out -


(sorry, screenshot is hard to get the photos just right, but the top line is basically what you want to watch for)

I will shut off the filter for now.

Next is PREDELAY. Predelay is the time it takes from the initial sound to the first reflection that you hear. (kind of like how far away you are from the relective object) This is really important to get just right on a reverb. Sometimes, if your reverb predelay is set too quickly, it can muddy the attack sound of a certain instrument. However, if it is too long, it could interfere with an upcoming element in drum loop or a synth lead. You must use this with caution, and find the right distance for the first reflection. If you are just looking for a light reverb sound, a short predelay would be fine....but, you can make some GREAT sounding reverb effects with a long predelay. Also, if you have the dry/wet knob up the whole way while this is running, you will ONLY hear the first reflection, and not the initial sound...which will create a nice delay depending on the predelay time.

For the purpose of understanding this tutorial, you can set your predelay to a LONG setting so that you can hear the first reflection distinctly, and therefor understand what the reverb by itself sounds like. It will sound more like a delay, but it will be distinct.

Next we have the EARLY REFLECTIONS section. This helps sculpt the sonic properties of the first reflections that you hear. Lets start with SHAPE. You can read your ableton manual for the exact science of what it is doing...but from a users standpoint...with a lower setting...the reverb will be a little muddier and longer. With a higher setting, the first reflections are lower in db, softer, and faster. The long reverb does not seem to hang around as long.

SPIN - adds a little bit of modulation to the early reflections and is one of the less strong parameters in the reverb (in my opinion...I guess you could argue) You can control the modulation frequency and depth with the x - y axis.

GLOBAL SETTINGS - Eco, mid, high - These are performance settings. HIGH uses up more CPU, and the lower settings perform better when running tons of other things.

Size - Controls the rooms size. This is a really neat parameter. First of all, I think this makes it feel like there are pipes in the room or like, really metal type walls when set low. Very metallic sounding. With a higher setting, you get a more shifted reverb sound...a little more natural. One really cool thing about this is if you automate from RIGHT to LEFT...you can make almost a flange sounding reverb if done quickly on a stab or snare. This is a really creative part of the reverb. Also, depending on where you start the size and automate it LEFT TO RIGHT, you can create a new tones, and then SLIDE it into a washed out reverb. If you were to find the same (or harmony) pitch of the sound being reverberated, you can get a nice effect.

STEREO - This is pretty straight forward. All the way to the right, the left and right channels are giving a different stereo reverb sound (natural when in rooms because your ears hear different sounds reflected off of different objects depending on which side the reflection came from). All the way set to the left, the reverb is mono. This sounds very FAKE and unnatural...but can be useful in production.

DIFFUSION NETWORK - These settings connect also to the DIFFUSE knob on the right side (which controls in db, the volume of diffusion). The diffusion is the decay of the early reflections. There are low & high shelf filters which can help you set the frequency of your reverb. These are very important also when making a reverb sound nice. It helps to match the frequency with the actual frequency of the instrument....or, shut it off if you want it natural. There are frequency and bandwidth options on this also. This is also a nice parameter to automated on a stab or snare to get a nice FILTER within the diffusion. With both of the HIGH & LOW deselected, the filters are off.

Now, those filters are VERY MUCH affected by the DECAY TIME. This part is SO important to get just right. It is basically HOW LONG your reverb will last. By setting this to the left, you can get a nice, short, small, reverb to help polish up a stale sound...or, set it to the right...and get a loud, overlapping reverb sound that will definitely annoy the hell out of the neighbors. However, with the decay time set all the way to the right, and some automation within compose...some nice sweeps and other weird hits can be made if later modulated with an LFO, and given some volume automation.

You can see in this picture, that with the decay time set to the maximum, a constant sound is heard and doesnt really drop too far in db really. I know its hard to tell because mac snapshot doesnt react fast enough, but in this picture, where the DARK bars are sitting, is the lowest they are going without actually dropping away like the normal snare hit does.

Freeze - This will freeze the diffused reverb when captured. This can get really annoying too as the sound will stay frozen until deselected. This has some creative uses in dance music. The cut, when on, makes it so if the reverb is activated again while frozen, it does not ADD more sound the the frozen reverb. If it is off, it will add on top of that reverb again and again as the reverb is activated.

FLAT - This is a really neat little parameter. It works in conjunction with your diffusion filter. If the FLAT is deselected, whatever frequency you have being changed in your filter...while frozen, will slowly lose energy in those frequency bands. If you could see a video of the spectrum, you could actually watch the high frequencies move slowly as they fade away. (cool effect) If it is on, those frequencies will remain the same.

Density and Scale control the diffusion just a little bit more. Sometimes when you play with these parameters, you dont notice much of a change. But, if you go to the SIZE knob, and make your room size really small, put your density up, and then move the SCALE knob...you will notice a big change. You can even automat the scale knob with a small room size for a nice flanger style effect to your reverb on a short hit. But when moving the scale, you can also change the diffusions pitch a little, so when using this in its creative form while making a track, be sure to check the pitch of the scale setting, especially if your reverb room size is small.

CHORUS - This is pretty straight up. Adds chorus to the diffusion...basic modulation when 2 sounds are played a little bit out of phase. You can also control the db level and frequency...same as a regular chorus x - y axis.

All of these are then controlled by the 3 knobs on the right hand size - REFLECT, DIFFUSE, DRY/WET.

Reflect is the reflecting level in db. You can make it louder, or softer.

Diffuse is the diffusion level in db. You can make it louder, or softer.

Dry / Wet is how much of the source signal and how much of the effect signal is being heard. To the left, no effect...only source signal. To the right, only effected sound, no source sound.

Well, that just about covers everything. Hope this helped. Make sure to shout what you know about reverb here, and make it echo!

Peace!

FroBot

What is Sound? The Physics of Sound.

Before I write this, please realize, that I am not an expert. I am just a music nut who studies all parts of sound, and then write about everything I learn to help my brain remember everything. I did not go to school for this, but have been self studying for years. This is just a basic scientific understanding of sound, explained with graphs from all over the internet I have found to make my points stand out. Without people putting free info on the internet for me to study, I would never know any of this....so, I felt like I should give back by making a nice understanding with better graphs and in my own words. This is to help me, and if it helps you, great!

What is Sound?

A sound is a vibration. No vibration, no sound. When a source vibrates, it moves particles beside it and vibrates in a chain reaction. Usually, this means air particles, but it can also be water, gas, and other kinds of particles. But, here, we will talk about air particles because that is what we usually deal with in music. When a source vibrates, and then vibrates particles, it comes from a source location (where it is the strongest), and then moves outward in wave fashion until it reaches and vibrates our eardrums. The sound does not travel in bulk, but rather in a sound wave coming from the source, and then disturbs the surrounding area...slowly getting weaker. This is simple energy displacement. Sound requires a medium (air particles), therefor, sound cannot travel in a vacuum. It reacts the same way as if you threw a rock into a still lake...the water ripples will form, and then disappear...returning to its original form.


Now, once this displacement of particles has occurred, we are left with 2 regions. 1 region is of high particle density, and the other is of low particle density (or high pressure, low pressure).

2 keys words to remember

Compressions (Condensation) - Regions of high particle density
Rarefactions - Regions of low particle density.
Rarefactions and Compressions both move in the same direction that the wave travels.


Now, the particles of air don't actually MOVE (physically) in the direction of the wave movement, but instead move within their normal positions in a cycle fashion. 1 cycle is when a particle moves from its STARTING POSITION, to the maximum displacement distance in one direction, back to the starting point, and then to the maximum displacement distance in the other direction.

Now this is just one cycle. Particles can vibrate THOUSANDS of times per second in this fashion. The number of cycles completed by a wave in one second is considered the FREQUENCY of vibration. One way to easily tell how many cycles a wave is making is by listening to the pitch. The pitch is very much affected by the number of cycles occurring. (more cycles at higher pitches)

In music terms, 1 cycle = 1Hz. (Hertz), 1000Hz = 1kHz.

The human ear can hear (depending on the person of course) anywhere from 16Hz - 20000Hz
(normal being around 20Hz - 18,000)

A normal piano range is in the graph below -

But just because a wave is at a certain frequency (on an instrument), does not mean it is going to SOUND the same. This is because of TIMBRE of the sound. If you play C4 on a piano, it will sound different from a C4 on a flute or a tuning fork.

To understand this, you must understand PURE TONES and OVER TONES.

A pure tone is one that is created by a perfect wave (like a tuning fork, or a tone generator). These contain a ONE TONE frequency, meaning, it only creates one tone on the frequency spectrum. A note played on a piano or any other instrument will have OVER TONES, which are other tones within the spectrum accenting the fundamental tone....giving it a unique sound. The tone with the lowest frequency is called the FUNDAMENTAL TONE. All other tones are OVER TONES.

This is where we get into harmonics. If an over tone has a frequency value that is a multiple of the fundamental tone ( x2, x3, x4, up to x14), these tones are called HARMONICS. It is the difference of each instruments over tones and fundamental tones that make up the sounds TIMBRE. Therefor, even though an instrument may produce a fundamental tone at say 4kHz, it has overtones at much higher frequency values. This is why most DAWs and other recording gear must be able to handle a much higher frequency than can actually be produced by the fundamental tone of an instrument. There will be MUCH loss in the dynamics of the sound if it is not able to be captured. Some instruments like cymbals or dog whistles produce frequency levels much higher than the human ear can hear, but they effect the whole sounds TIMBRE if not captured correctly. This is why many analog purist stay analog, and not digital...where most frequencies are cut at 20,000Hz. Below is is a graph of a general saw wave and its harmonic overtones.



Now there are 2 types of sound waves (we will break these 2 kinds down even more in a minute)

One is a wave with a definite pitch (we call a note). Another kind is one with no definite pitch (we call a noise). Music has both almost all of the time. The difference is definitely audible, but what is the scientific difference? A note contains regular vibrations (periodic motion), and a noise contains irregular vibrations(non-periodic motion).



Now if we break down the periodic motion vibrations, some of the common vibrations are below -
Most of these waves are created by synthetic sounds. Traditional instruments function in a sine wave form.

In music, we divide these sound sections into groups called OCTAVES. 7 different notes A-G, & the 8th note being the start of the next octave. Each note named the same letter sounds very similar to the others named the same letter because they are in a multiple of the others frequency. (x. 220Hz, 440Hz, 880Hz). This means the actual frequency rate (or cycles) is 2ce the speed of the previous. These notes are said to differ in pitch by ONE octave.

Now lets talk a little about amplitude. A loud sound is produced by more violent vibrations than that of a weak sound. The amplitude is the maximum displacement of the air particles from their original place. As the sound wave continues to displace particles in a wave fashion, it is displacing energy (energy loss). That is why the sound will get weaker and weaker. The energy is displaced in the form of HEAT, and the sound will get weaker as it travels away from its source. To understand graphically, the farther away from the center line on the graph that the wave gets, the higher in amplitude it is.

Now scientifically, the amount of energy loss follows the "Inverse Square Law". (but keep in mind, this is not what the human ear actually hears...because of our ears compensation system).

Scientifically, if the distance from the source is TWICE the distance, the sound drops to one quarter. If the distance is 4 times the distance, it drops to one-sixteenth.


Now, if you think about this....this doesn't make much sense. Because if you are at a club, listening to some speakers, and you move twice your the distance away from the speakers, the audio does not seem to drop in these proportions. This is because of the compensation system in our ears due to the 3 pivot system in place. It helps your ear create leverage on desired frequencies. This system lets your ears hear weaker sounds at maximum strength, while louder sounds are reduced to prevent damage to your ears. That means, that the scientific measurable amplitude differs from what is heard, and what is actually happening. We refer to this phenomenon as "LOUDNESS". Loudness is what we use to measure what our EAR hears in change, and not the scientific term of amplitude. Loudness is also determined by how sensitive your ears are, and differs person to person. It is measured in decibels. (db)

Now, what happens when 2 sound sources meet at the same time. If 2 waves arrive at the same place at the same time, at the same high pressure and low pressure displacement areas...we will get what is called "Reinforcement". (in-phase) This will result in a greater wave intensity. If 2 waves arrive, and one is a half of a cycle late, the high pressure of the one wave, and the low pressure of the other wave will cancel each other out. This is what is called "cancellation". (out of phase)

In the photo, graph 1 is in phase, graph 2 is out of phase.

If we were to set up 2 speakers producing perfect waves at 1/2 the cycle speed difference, this is how it would sound to your ear (cancelled) -



Now, how do we measure all of this. Well, the equation for "Speed of Sound" is rather simple. I liked this tattoo to help explain -


V = speed of sound
f = frequency
λ = wavelength (lambda in greek)

A wave front is a surface on which all particles are in the same phase of vibration. The distance between the 2 wave fronts is called the wavelength. (or the distance between 2 separate compressions, or 2 separate rarefactions)

Now obviously, the speed of sound is affected by other things. This equation only works in a perfect environment with no obstacles. In an everyday situation, we have air pressure, and physical objects that can obstruct a waves path.

When a sound wave is made in air thats temperature changes with altitude, REFRACTIONS occur. This is why sound can travel at different speeds depending on air pressure and particle density.



So obviously the equation for the speed of sound has variables.


Well that's it for my small piece about sound. If you have a better understanding of how sound works, you can begin to change it the way you want rather than just randomly.

Peace!

FroBot